- Original Message
From: Andy Armstrong <[EMAIL PROTECTED]>
> Can you confirm that this works OK with your runtests?
Yup. Works fine. It's me not forcing the shared memory value to be a number,
so it looks like we had strange behavior due to padded NULLs. Not sure if this
should be
- Original Message
From: Andy Armstrong <[EMAIL PROTECTED]>
> Taking out the shared memory calls and setting the test number
> explicitly makes it work under runtests for me.
Crap. I get the same results. I'm suspecting that the Test::Class problem is
unrelated to this one (though I w
Ovid wrote:
> I can never get it to work with the Test::Class code, using either prove or
> runtests. In the regular .t style program, it succeeds if I use a leading
> plan, but fails if I use a trailing plan with runtests:
>
> use Test::More tests => 3;
>
> Versus:
>
> use Test::More 'no
- Original Message
From: Andy Armstrong <[EMAIL PROTECTED]>
> > Parse errors: No plan found in TAP output
>
> D'oh - sorry - should have read the output from your runtests properly :)
>
> So we're losing the plan somehow.
I can never get it to work with the Test::Class code, using eith
- Original Message
From: Andy Armstrong <[EMAIL PROTECTED]>
> Interestingly runtests is happy for me:
>
> [13:38] andy $ runtests -v forkingtest.pl
> forkingtest...ok 1 - mytest called
> ok 2 - in the child
> ok 3 - in the parent
> 1..3
> All 3 subtests passed
>
> Test Summary Repor
Andy Armstrong wrote:
> Test Summary Report
> ---
> forkingtest.pl (Wstat: 0 Tests: 3 Failed: 0)
> Parse errors: No plan found in TAP output
> Files=1, Tests=3, 0 wallclock secs ( 0.03 cusr + 0.02 csys = 0.05 CPU)
D'oh - sorry - should have read the output from your runtests p
Ovid wrote:
> - Original Message
> From: Andy Armstrong <[EMAIL PROTECTED]>
>
>> What happens if you explicitly set the test number to what it should be?
>
>>$test->current_test( 3 ); # < hard wired
>
> Exact same behavior, which implies (to me) that my use of s
Ovid wrote:
> Basically, if I fork a child, when we return, Test::Builder will not have the
> new test count, so I explicitly use shared memory [1]. While the above runs
> just fine with prove, runtests (from the TAP::Parser distribution), fails
> with the following:
>
> ok 1 - mytest called
- Original Message
From: Andy Armstrong <[EMAIL PROTECTED]>
> What happens if you explicitly set the test number to what it should be?
>$test->current_test( 3 ); # < hard wired
Exact same behavior, which implies (to me) that my use of shared memory is
correct.
Ovid wrote:
> Am I doing something really stupid here? Could I be using shared
memory incorrectly?
What happens if you explicitly set the test number to what it should be?
if ( !defined( my $child_pid = fork() ) ) {
diag "Cannot fork: $!";
}
elsif ( !$child_pid ) {
ok
10 matches
Mail list logo
